Scholarlink

<p>Outliers can significantly distort statistical inference, leading to unreliable estimations. This paper proposes a novel sampling design, Robust Except Extremes Ranked Set Sampling (<i>REERSS</i>), which enhances estimation robustness by excluding extreme observations from each set. The <i>REERSS</i> mean estimator is shown to be unbiased under symmetric distributions and outperforms traditional methods such as Simple Random Sampling (<i>SRS</i>), Ranked Set Sampling (<i>RSS</i>), and <i>RSS</i> with Extremes (<i>RSSE</i>). Monte Carlo simulations and real-world applications, including datasets from the biomedical domain, demonstrate the superior efficiency and accuracy of <i>REERSS</i> in mean estimation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Robust Except Extreme Ranked Set Sampling (REERSS): An Approach to Outlier Mitigation

  • 期刊论文

  • 发表日期：

    2026年4月9日

查看全文
  • R. R. Sinha,
  • Alok Kumar

摘要

Outliers can significantly distort statistical inference, leading to unreliable estimations. This paper proposes a novel sampling design, Robust Except Extremes Ranked Set Sampling (REERSS), which enhances estimation robustness by excluding extreme observations from each set. The REERSS mean estimator is shown to be unbiased under symmetric distributions and outperforms traditional methods such as Simple Random Sampling (SRS), Ranked Set Sampling (RSS), and RSS with Extremes (RSSE). Monte Carlo simulations and real-world applications, including datasets from the biomedical domain, demonstrate the superior efficiency and accuracy of REERSS in mean estimation.